Accelerated failure time models for the analysis of competing risks

Sangbum Choi, Hyunsoon Cho

Research output: Contribution to journalArticle

Abstract

Competing risks are common in clinical cancer research, as patients are subject to multiple potential failure outcomes, such as death from the cancer itself or from complications arising from the disease. In the analysis of competing risks, several regression methods are available for the evaluation of the relationship between covariates and cause-specific failures, many of which are based on Cox's proportional hazards model. Although a great deal of research has been conducted on estimating competing risks, less attention has been devoted to linear regression modeling, which is often referred to as the accelerated failure time (AFT) model in survival literature. In this article, we address the use and interpretation of linear regression analysis with regard to the competing risks problem. We introduce two types of AFT modeling framework, where the influence of a covariate can be evaluated in relation to either a cause-specific hazard function, referred to as cause-specific AFT (CS-AFT) modeling in this study, or the cumulative incidence function of a particular failure type, referred to as crude-risk AFT (CR-AFT) modeling. Simulation studies illustrate that, as in hazard-based competing risks analysis, these two models can produce substantially different effects, depending on the relationship between the covariates and both the failure type of principal interest and competing failure types. We apply the AFT methods to data from non-Hodgkin lymphoma patients, where the dataset is characterized by two competing events, disease relapse and death without relapse, and non-proportionality. We demonstrate how the data can be analyzed and interpreted, using linear competing risks regression models.

Original languageEnglish
JournalJournal of the Korean Statistical Society
DOIs
Publication statusAccepted/In press - 2018 Jan 1

Fingerprint

Accelerated Failure Time Model
Competing Risks
Failure Time
Covariates
Linear regression
Modeling
Cancer
Cause-specific Hazard
Cumulative Incidence Function
Non-Hodgkin's Lymphoma
Competing Risks Model
Cox Proportional Hazards Model
Hazard Function
Risk Analysis
Complications
Regression Analysis
Hazard
Regression Model
Regression
Simulation Study

Keywords

  • Accelerated life testing
  • Cause-specific hazard
  • Cumulative incidence
  • Inverse probability weighting
  • Linear regression
  • Survival analysis

ASJC Scopus subject areas

  • Statistics and Probability

Cite this

Accelerated failure time models for the analysis of competing risks. / Choi, Sangbum; Cho, Hyunsoon.

In: Journal of the Korean Statistical Society, 01.01.2018.

Research output: Contribution to journalArticle

@article{f588eed9a2944d778e0b70514c8d1675,
title = "Accelerated failure time models for the analysis of competing risks",
abstract = "Competing risks are common in clinical cancer research, as patients are subject to multiple potential failure outcomes, such as death from the cancer itself or from complications arising from the disease. In the analysis of competing risks, several regression methods are available for the evaluation of the relationship between covariates and cause-specific failures, many of which are based on Cox's proportional hazards model. Although a great deal of research has been conducted on estimating competing risks, less attention has been devoted to linear regression modeling, which is often referred to as the accelerated failure time (AFT) model in survival literature. In this article, we address the use and interpretation of linear regression analysis with regard to the competing risks problem. We introduce two types of AFT modeling framework, where the influence of a covariate can be evaluated in relation to either a cause-specific hazard function, referred to as cause-specific AFT (CS-AFT) modeling in this study, or the cumulative incidence function of a particular failure type, referred to as crude-risk AFT (CR-AFT) modeling. Simulation studies illustrate that, as in hazard-based competing risks analysis, these two models can produce substantially different effects, depending on the relationship between the covariates and both the failure type of principal interest and competing failure types. We apply the AFT methods to data from non-Hodgkin lymphoma patients, where the dataset is characterized by two competing events, disease relapse and death without relapse, and non-proportionality. We demonstrate how the data can be analyzed and interpreted, using linear competing risks regression models.",
keywords = "Accelerated life testing, Cause-specific hazard, Cumulative incidence, Inverse probability weighting, Linear regression, Survival analysis",
author = "Sangbum Choi and Hyunsoon Cho",
year = "2018",
month = "1",
day = "1",
doi = "10.1016/j.jkss.2018.10.003",
language = "English",
journal = "Journal of the Korean Statistical Society",
issn = "1226-3192",
publisher = "Korean Statistical Society",

}

TY - JOUR

T1 - Accelerated failure time models for the analysis of competing risks

AU - Choi, Sangbum

AU - Cho, Hyunsoon

PY - 2018/1/1

Y1 - 2018/1/1

N2 - Competing risks are common in clinical cancer research, as patients are subject to multiple potential failure outcomes, such as death from the cancer itself or from complications arising from the disease. In the analysis of competing risks, several regression methods are available for the evaluation of the relationship between covariates and cause-specific failures, many of which are based on Cox's proportional hazards model. Although a great deal of research has been conducted on estimating competing risks, less attention has been devoted to linear regression modeling, which is often referred to as the accelerated failure time (AFT) model in survival literature. In this article, we address the use and interpretation of linear regression analysis with regard to the competing risks problem. We introduce two types of AFT modeling framework, where the influence of a covariate can be evaluated in relation to either a cause-specific hazard function, referred to as cause-specific AFT (CS-AFT) modeling in this study, or the cumulative incidence function of a particular failure type, referred to as crude-risk AFT (CR-AFT) modeling. Simulation studies illustrate that, as in hazard-based competing risks analysis, these two models can produce substantially different effects, depending on the relationship between the covariates and both the failure type of principal interest and competing failure types. We apply the AFT methods to data from non-Hodgkin lymphoma patients, where the dataset is characterized by two competing events, disease relapse and death without relapse, and non-proportionality. We demonstrate how the data can be analyzed and interpreted, using linear competing risks regression models.

AB - Competing risks are common in clinical cancer research, as patients are subject to multiple potential failure outcomes, such as death from the cancer itself or from complications arising from the disease. In the analysis of competing risks, several regression methods are available for the evaluation of the relationship between covariates and cause-specific failures, many of which are based on Cox's proportional hazards model. Although a great deal of research has been conducted on estimating competing risks, less attention has been devoted to linear regression modeling, which is often referred to as the accelerated failure time (AFT) model in survival literature. In this article, we address the use and interpretation of linear regression analysis with regard to the competing risks problem. We introduce two types of AFT modeling framework, where the influence of a covariate can be evaluated in relation to either a cause-specific hazard function, referred to as cause-specific AFT (CS-AFT) modeling in this study, or the cumulative incidence function of a particular failure type, referred to as crude-risk AFT (CR-AFT) modeling. Simulation studies illustrate that, as in hazard-based competing risks analysis, these two models can produce substantially different effects, depending on the relationship between the covariates and both the failure type of principal interest and competing failure types. We apply the AFT methods to data from non-Hodgkin lymphoma patients, where the dataset is characterized by two competing events, disease relapse and death without relapse, and non-proportionality. We demonstrate how the data can be analyzed and interpreted, using linear competing risks regression models.

KW - Accelerated life testing

KW - Cause-specific hazard

KW - Cumulative incidence

KW - Inverse probability weighting

KW - Linear regression

KW - Survival analysis

UR - http://www.scopus.com/inward/record.url?scp=85057067835&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85057067835&partnerID=8YFLogxK

U2 - 10.1016/j.jkss.2018.10.003

DO - 10.1016/j.jkss.2018.10.003

M3 - Article

AN - SCOPUS:85057067835

JO - Journal of the Korean Statistical Society

JF - Journal of the Korean Statistical Society

SN - 1226-3192

ER -